Merge pull request #262 from Eliyahu-Ravuna/patch-10

Access memory predictably - Revised according to the feedback in #256
This commit is contained in:
Gabriel Dos Reis 2015-10-02 06:40:59 -07:00
commit f72217f30c

View File

@ -8660,7 +8660,19 @@ make the job of the optimizer much harder. Simple code often optimizes better th
Performance is very sensitive to cache performance and cache algorithms favor simple (usually linear) access to adjacent data.
???
##### Example
int matrix[rows][cols];
//bad
for(int c=0; c<cols; ++c)
for(int r=0; r<rows; ++r)
sum += matrix[r][c];
//good
for(int r=0; r<rows; ++r)
for(int c=0; c<cols; ++c)
sum += matrix[r][c];
### <a name="Rper-context"></a> PER.30: Avoid context switches on the critical path